virgildotcodes

6 ms·

1.
by virgildotcodes 7d ago
I agree. It won’t stop us from othering the source of our demise as quickly as we are able. There’s also the argument to be made that your child is their own person, and you only have so much responsibility/control over their actions.
2.
by virgildotcodes 7d ago
At least we’ll eventually have an entity other than ourselves to blame for our annihilation.
3.
'This is crazy. This is insane': Bill Gates has changed his mind about AI, jobs(semafor.com)
5 points by virgildotcodes 21d ago|2 comments
4.
by virgildotcodes 26d ago
Does this mean a commensurate increase in subscription usage limits?
5.
by virgildotcodes 29d ago
Hm, seems to be a bit of a contradiction between what you’re saying in terms of “insurers want to pay more” and the infamous tendency of insurers to deny claims as much as possible?
6.
by virgildotcodes 1mo ago
If you read up on the sequence of events posted above you’ll see that the admin exhausted every avenue available to them but was ultimately defeated in their efforts here.
7.
by virgildotcodes 1mo ago
Why are we even playing games about implications when we can see the empirical fact that the admin’s policies on Palestine over the last years have been to carpet bomb people?
8.
by virgildotcodes 1mo ago
This apologia loses force when you can transparently see that Ben Gvir’s definition of terrorist and Palestinian are a single circle Venn diagram.
9.
by virgildotcodes 1mo ago
The funny thing is it doesn’t even code well. It’s still worse than OpenAI, Anthropic, and the leading open source models. So you get worse outcomes while also financially supporting one of the most prolific pushers of race wars, the destru
10.
by virgildotcodes 1mo ago
They may need approval from their employers.
11.
by virgildotcodes 1mo ago
I'm sold.
12.
by virgildotcodes 1mo ago
OpenAI and Anthropic need to just go ahead and give people access to the cyber models. Otherwise we have a world of attackers using open and closed source models against a much smaller group of maintainers that are likely heavily dependent
13.
by virgildotcodes 1mo ago
Not that useful when large swathes of the population are unemployed, which is where this is all aiming.
14.
by virgildotcodes 1mo ago
The project was originally called clawdbot as it was based on Claude. Pressure from Anthropic made Peter change the name to Moltbot, which everyone thought sounded gross, so he changed it again to openclaw, with a lobster mascot.
15.
by virgildotcodes 1mo ago
This is basically openclaw with browser access, or am I missing something?
16.
by virgildotcodes 1mo ago
The admin fought quite hard to destroy the electric conversion, not just cancelling future orders but scrapping existing trucks and infrastructure despite that being a major expense. It's in line with the way it's fighting against
17.
by virgildotcodes 1mo ago
Wow, a small glimmer of good news. I almost want to hide this comment over fear of drawing attention to this thing that got away from the powers that be.
18.
by virgildotcodes 1mo ago
Is the game you're playing "see if I can be subtle enough to get away with undermining renewables through misleading statements on social media without people calling me out on it"? If so I think you're down 1 - 0 so far
19.
by virgildotcodes 1mo ago
I recently saw some discussion about the CPC for personal injury lawyers being ~$200 on Google ads. Seems insane to me, but I’m totally disconnected from the advertising world. That said, depending on the conversion rates between chats and
20.
The Maxwell conjecture is false. Counterexample found by AI (GPT-5.6 Sol)(xcancel.com)
3 points by virgildotcodes 2mo ago|2 comments
21.
Apple Upgrade Launches in the United States(apple.com)
12 points by virgildotcodes 2mo ago|4 comments
22.
by virgildotcodes 2mo ago
It's likely bumping up against people's desire to have the model complete a given task without asking the person to intervene a bunch of times. Seems unclear how you satisfy everyone here.
23.
by virgildotcodes 2mo ago
> you could do the same back then with multiple models and more targeted finetuning I mean, come on, this is just not true. You could not achieve anything like what you can with modern agentic coding with Fable / 5.6 Sol from any co
24.
Dinitz-Garg-Goemans conjecture proven false by GPT 5.6(twitter.com)
5 points by virgildotcodes 2mo ago|0 comments
25.
by virgildotcodes 2mo ago
I think the ultimate goal here is to protect the valuations of Anthropic, OpenAI, xAI etc. so probably?
26.
Trump administration reportedly reviving push to ban Chinese AI models, Kimi K3(tomshardware.com)
4 points by virgildotcodes 2mo ago|5 comments
27.
by virgildotcodes 2mo ago
When you eat alphabet soup, do you take note of all the leftist messages bobbing in and out of existence in your bowl?
28.
by virgildotcodes 2mo ago
> how much users want to use/pay for it, which is a pretty fuzzy problem Isn’t this quantifiable by revenue?
29.
by virgildotcodes 2mo ago
Proof linked here - https://cdn.openai.com/pdf/04d1d1e4-bc75-476a-97cf-49055cd98... Prompt used - https://cdn.openai.com/pdf/04d1d1e4-bc75-476a-97cf-49055cd98... Lean formalization of the proof -
30.
Cycle Double Cover Conjecture Proof Using 5.6 Sol Ultra(twitter.com)
1 points by virgildotcodes 2mo ago|2 comments

More ›